Considering knowledge gaps for automated driving in conventional traffic

Abstract

—With increasing numbers of vehicles using low level automation and higher level automation expected in the future, significant effects are expected on traffic flow. Despite much simulation and driving simulator research on SAE level 1 vehicles, there remain many questions in regard to the effects of the systems on traffic flow. The effects of higher levels of automation are even more difficult to estimate, as these vehicles are not even present on roads at this time, let alone in sufficient numbers to analyze. In this research, we propose a methodology for a-priori analysis of potential conflict situations: Method for Explorative TRaffic scenario Observation and Analysis (METRO-A). It is applied to the case of automated driving in conventional traffic to analyze potential difficulties that SAE level 3 and 4 and higher vehicles may encounter in mixed traffic conditions, for a weaving section case-study. Furthermore, a set of important research questions are constructed that are relevant for the automotive industry, and road agencies and authoritie